Page 179 - Chalcogenide Glasses for Infrared Optics

Glass Pr ocesses for Other Applications 155

glass was to have an index difference from the core glass of –0.01 to −0.03
to provide the desired acceptance angle or numerical aperture (NA).
The NA may be calculated from (n 2 – n 2 ) 1/2 = sin α/2 where n is the
core clad
index and α is the total included acceptance angle. All the optical fiber
3
formulas are derived from Snell’s law: N sin θ = N sin θ , where N and
1 1 2 2 1
N are the indices in medium 1 (air in this case) and medium 2 (fiber
2
core); θ is the angle of incidence to the surface of the core and θ is the
1 2
angle of the refracted ray to the axis of the core. Part of the incident ray
is reflected from the surface of the core and lost while part is refracted
into the core. When the core ray strikes the interface with the clad glass,
the situation changes since the clad glass index is less than that of the
core index and the ray angle with the normal to the interface decreases
to a critical angle θ where the ray is totally reflected internally. The angle
C
−1
3
may be calculated from θ = cos (N /N ). The refracted ray then travels
C 2 1
down the core with little or no reflection loss. However, when a ray
enters the core at a greater oblique angle (outside the NA) such that the
refracted ray strikes the core-clad interface at an angle larger than θ , the
C
ray is poorly reflected, partially absorbed in the cladding, and finally
disappears.
For laser light transmission through a fiber, a small acceptance
angle is desirable with the beam focused using a long-focal-length
lens or mirror to a spot size smaller than the core diameter of the fiber.
The rays then enter the fiber at angles close to normal and are refracted
still closer to normal in the fiber. Loss of energy is minimized through
fewer internal reflections off the walls of the fiber. Expansion of the
clad glass was to be 95 percent of the core glass to minimize stress
between core and clad. After the goals of the program were realized
and production was started, Codman would guarantee AMI substan-
tial sales of glass to Codman for 3 years.
Glass systems as candidates for producing a glass meeting the goals,
in order of probability, were (1) As-Se-Te, (2) Ge-As-Se-Te, and
(3) Ge-As-Se glasses based on Amtir 1 diluted with tellurium. Amtir 1
glasses low in germanium and As-Se-Te glasses were all evaluated rela-
4
tive to lowest absorption level at 10.6 µm. Results supported the best
glass would be found in the As-Se-Te system. Considerable information
concerning these glasses was found in the work of Joe Jerger in the Servo
report. A number of glass compositions were made and tested. They all
5
seemed to have absorption levels at 10.6 µm related to the amount of
selenium in their composition. That fact indicated that either an impurity
in most all selenium or an intrinsic selenium-selenium vibration over-
tone was the cause of the limiting absorption level. AMI purchased high-
purity selenium from suppliers in Japan, Canada, Germany, Belgium,
France, and KBI and Asarco in the United States. Comparison of furnished
analysis showed major impurities of Cu, Fe, Si, As, Te, and S. Virtually all
selenium is produced as a by-product of copper production. The selenium
is chemically separated and then purified by physical distillation.